KDnuggets : News : 2008 : n08 : item6 | PREVIOUS | NEXT |
FeaturesFrom: Charles ElkanDate: 18 Apr 2008 Subject: Fifth Annual UC San Diego Data Mining Contest
Sponsored by Fair Isaac UC San Diego and Fair Isaac are pleased to announce the Fifth Annual University of California San Diego (UCSD) Data Mining Contest. The contest consists of two classification tasks based on data from a physics experiment.
Contest Tasks Task 1 (fully-labelled, supervised): This is a standard supervised binary classification task, with a test set of examples consisting of both positively- and negatively-labelled examples. Performance on this task is evaluated by classification accuracy on the test set. Task 2 (positive-only, semi-supervised): This is also a binary classification task, but only a few of the examples in the test set are labelled (all positive). The unlabelled examples are both positive and negative. Performance on this task is evaluated by F1 score on the test set.
Rules Once a day, teams may estimate their performance by submitting their predictions on a quiz set that is of similar construction to the test set. The scores on this set are displayed on the contest leaderboard.
Prizes $8000 in prizes will be awarded based on task and division ($2000 per task per division):
First Place $1000 Undergraduate and graduate teams will be judged separately. Performance on the two tasks will also be judged separately.
Schedule: Final Submissions Due: June 15, 2008. |
KDnuggets : News : 2008 : n08 : item6 | PREVIOUS | NEXT |
Copyright © 2008 KDnuggets. Subscribe to KDnuggets News!